Abstract:
Abstract: Solar energy is one of the most potential renewable sources in most rural areas. Photovoltaic (PV) power generation technology has also been widely applied with abundant solar energy resources. The installed capacity of PV power generation is ever increasing in recent years in China. However, the excess or abandonment of the output has often occurred, due mainly to the mismatch between the total load and energy output curve of PV generation. An optimal control strategy is highly required to absorb the excess PV power for the cost-saving system operation. The regional PV absorption characteristics also vary greatly in some areas, particularly where the traditional time-of-use (TOU) electricity price cannot be suitable for the local load. The incentive demand response (IDR) can be expected to serve as the scheduling strategy from the demand side, according to subsidies and discounts for the flexible load. In this study, a compound strategy of demand side response was proposed for the distribution networks with a high proportion of the distributed PV using the combined IDR and regional centralized regulation (RCR). Among them, one hour was taken as the time interval. The specific control strategy was designed as follows. Firstly, the peak and valley periods of electricity price were divided in the regional dispatching center (such as the load aggregator or the power selling company) using the IDR and original TOU electricity price in the study area. A new TOU electricity price scheme was obtained, according to the local PV output and load time sequence. The difference in electricity price before and after adjustment was selected as the subsidy unit price in the 24 control periods, and then transmitted to the power load in advance. Secondly, the distributed model was optimized by the load user. A power consumption model was then achieved using the user load reference value and the new TOU electricity price. The range of the distributed load was determined to match the power consumption plan and load adjustment range of each period of the day in the regional dispatching center. Thirdly, the PV power prediction was combined with the load plan uploaded by the user and the adjustment range. A centralized control model was established with the distributed calculation as the variables, while the maximum PV consumption and the lowest economic operation cost as the objective. Then, the optimal plan was obtained for the load adjustment. The user load was controlled using the direct load control (DLC) on the control day. A better match was realized for the curve of the load and PV output, in order to maximize the PV consumption for the overall economic operation of the system. Taking a 10 kV distribution network as an example, the model was verified using the MATLAB software. As a result, the total cost of one day after optimization using the compound demand response strategy was reduced by 12.9% and 10.5%, respectively, compared with the traditional TOU electricity price and the user IDR with the new TOU electricity price only without RCR. Consequently, the control strategy of decentralized and regional centralized dispatching can be expected to match the total load curve with the PV output curve for the interaction between the supply and demand, according to the response income of the user demand side. At the same time, the finding can greatly contribute to the consumption capacity of PV power generation and the economy of system operation